Build a restaurant database from live Google Maps
A fixed restaurant database goes stale fast. Pull fresh records straight from Google Maps instead: name, phone, website, hours, rating, and cuisine for any city, as clean JSON.
One request, real response
That exact call came back in about a second with 5 complete records. Set limit up to 100 restaurants per search, and add stars=true for each spot's 1-to-5 review histogram.
The full listing, ready for a database
Every record comes complete. No field tiers, no extra charge for hours or ratings:
Where Google shows them, records also carry the listing's attribute lists: service options (delivery, takeout, outdoor seating), dining options (breakfast through dessert), amenities, and payment types. The catch: Google Maps listings never include an email address, so these records don't either. The section below covers how to get a verified email for the owner.
From one search to the whole city
One search returns up to 100 restaurants. A city holds more than that, so CrustAPI sweeps the whole metro and hands back every restaurant, deduplicated for you, past Google's ~100-per-search cap. The result is a citywide restaurant table with no work on your side.
Two billing notes worth knowing up front. A search that comes back empty costs nothing, so trying one costs you nothing to run. You pay per record returned, so keep your searches focused if you are watching credits.
Three jobs it does well
| Who | What they build | The fields that matter |
|---|---|---|
| Food suppliers and distributors | A call list for every restaurant in the delivery zone | Phone, address, categories, and hours (so reps call between rushes) |
| POS and restaurant SaaS sales | Territory lists ranked by how busy a spot looks | Review count as a traffic proxy, rating, website, plus a verified owner email from LinkedIn people search |
| Analysts and researchers | Market maps: cuisine mix, density, and rating spread by area | Categories, lat/lng, rating, and the permanently-closed flag |
One credit per restaurant
The free tier gives you 3,000 credits a month with no card, so a 3,000-row restaurant table costs $0. Past that, prepaid packs: 25k for $49 · 100k for $149 · 500k for $549 · 2.5M for $1,999, ex-tax, all self-serve. That works out to $1.96 per 1,000 restaurants at entry. Credits never expire, so whatever a project doesn't use waits for the next one. And you're only charged when a query returns results; a search that returns 80 restaurants costs 80 credits, a search that returns none costs nothing.
Add the reviews, then find the owner
Swap type=maps for type=reviews to pull the actual Google reviews for any restaurant in your table: rating, text, author, and date, sorted and paginated, in about 0.3 seconds per call.
And when the job is outreach, the maps record gets you the restaurant while the LinkedIn people search endpoint gets you the person. It searches public LinkedIn profiles and returns a verified email on people results, which is the piece a Maps listing can never give you.
Related: Google Maps Scraper API · Google Reviews API · LinkedIn People Search API
When something else is the better pick: if you need a prebuilt national list today and don't want to run any queries at all, a dataset vendor's ready-made CSV gets you started faster, stale rows and all. And if you're building live map features inside an app (place lookups, autocomplete on a map view), Google's official Places API with its SLA is built for exactly that; here's how they compare. This page is for people who want fresh records they own and can keep.
Common questions
Clay
Start with one city
3,000 free credits covers a real citywide restaurant list. Run one search and look at the JSON that comes back.
Get 3,000 free credits